Follicular Lymphoma is the second most common Non-Hodgkin Lymphoma, with roughly 14,000 patients diagnosed annually in the US. Innovative treatments for follicular lymphoma using targeted monoclonal antibodies have improved the prognosis for patients, but these treatments are expensive. To the extent that these therapies can be targeted to patients in whom the disease is progressing, it will maximize the cost- effectiveness of the therapy. Early detection of disease progression is especially important in patients who have undergone therapy to induce remission, as additional rounds of therapy have the greatest impact when tumor burden is low. Unfortunately, diagnostic methods to monitor low level disease have not progressed as dramatically as treatments for follicular lymphoma, so there is a need to develop diagnostic tools that are broadly applicable to follicular lymphoma patients, and highly sensitive in detecting residual disease. We have developed methods for sequencing the somatically rearranged complementarity determining region 3 (CDR3) segments of immunoglobulin genes from large populations of B-cells. These CDR3 sequences can be used as biomarkers for clonal populations of B-cells, such as lymphoma tumors. In Aim 1 and 2, we propose to apply this technology to measure residual tumor burden in bone marrow specimens from follicular lymphoma patients undergoing chemotherapy for follicular lymphoma, and to assess the predictive value of these data for progression free survival. Our methods are more sensitive than PET scans, and should be applicable to significantly more than the half of patients who can currently be monitored by PCR of the IGH/BCL2 translocation. If our diagnostic methods can predict disease progression before it was detected in the clinical trials analyzed, then it might be possible to use our diagnostic to optimize monoclonal antibody therapies for patients who are at increased risk of disease progression. Aim 3 extends our analysis to peripheral blood samples drawn from the same patients, in the hope that residual disease might be detected in biospecimens that are much less invasive than bone marrow biopsies. If our methods can monitor disease burden in blood, this would facilitate much more frequent monitoring of disease state, and thereby allow for significantly earlier intervention in disease progressors than is currently possible. The proposed diagnostic methods have the potential to significantly improve therapeutic outcomes for follicular lymphoma, by focusing therapeutic intervention on patients at greatest risk of relapse, while increasing the effectivenes of treatment by detecting relapse earlier than is presently possible.